Axial motor control board, method, computer device and storage medium

Abstract

The application relates to an axial motor control board, a method, a computer device and a storage medium. The application sets a double processing architecture of a main control chip and a coprocessor in the axial motor control board. The main control chip is responsible for system scheduling, fault diagnosis and data management, so that the control board can still maintain stability under multi-task operation. The coprocessor is dedicated to executing high real-time tasks such as current decoupling, matrix calculation and coordinate transformation in the field-oriented control algorithm, so as to improve the dynamic response performance of the current loop and the speed loop. Precise motor state sensing and closed-loop control can be realized, and the control instruction transmission and operation data storage are more reliable through unified communication framework cooperation, so that the control precision, response speed and operation stability of the axial motor are effectively improved, and the reliability and safety of the control system under complex working conditions are enhanced.

Description

Technical Field

[0001] This application relates to the field of motor control technology, and in particular to an axial motor control board, method, computer equipment, and storage medium. Background Technology

[0002] In fields such as humanoid robots, high-end medical equipment, and new energy vehicles, axial motors have become core drive components due to their advantages such as high power density, flat structure, and lightweight design. However, the unique magnetic field distribution (magnetic flux distributed along the axial direction), high torque density requirements, and compact installation space of axial motors place significantly higher technical demands on their control boards than on traditional radial motor control boards. Current traditional control boards suffer from the following key problems:

[0003] First, insufficient control precision and real-time performance: Traditional control boards mostly use a single-chip architecture with limited processing power. When running the field-oriented control (FOC) algorithm, it cannot identify motor parameters in real time. For example, temperature increases cause winding resistance to rise by 10%-20%, and load changes cause inductance fluctuations, resulting in a decrease in current decoupling accuracy. The motor torque ripple rate increases from 5%-8% to 12%-15%, which cannot meet the sub-millimeter level control requirements of humanoid robot joints. Moreover, relying on traditional Ethernet communication results in a command response delay of 3-5 milliseconds, which can easily lead to motion lag in scenarios such as rapid robot grasping and precision assembly.

[0004] Second, the fault diagnosis and self-protection capabilities are weak: the existing control board only judges the fault by monitoring the current and voltage thresholds, with a misdiagnosis rate as high as 30%-40%, and cannot distinguish specific problems such as "load change", "winding short circuit" and "controller failure"; moreover, there is no fast protection mechanism when the fault occurs, which can easily lead to motor insulation aging and permanent magnet demagnetization, which may cause safety accidents, especially in scenarios such as medical surgical robots.

[0005] Third, poor heat dissipation performance adaptability: When the axial motor operates at high power density, the heat generated is concentrated. However, the traditional control board uses a common FR4 substrate and simple patch heat dissipation, which has low thermal conductivity. Under high temperature conditions (such as continuous high load conditions of industrial robots), the temperature of the core chip of the control board is easy to exceed 85°C, which leads to a decrease in the stability of the algorithm operation and even triggers overheat protection shutdown.

[0006] Fourth, low integration and compatibility: The interfaces of traditional control boards and axial motor encoders (such as absolute Gray code encoders) and brakes (electromagnetic brakes) are not uniform, requiring additional adapter modules, which increases the axial dimension by 15%-20% in volume and cost; and it is not compatible with axial motors of different power levels. When replacing the motor, the control board hardware needs to be redesigned, and the adaptation cycle is long, usually 2-3 months.

[0007] Fifth, insufficient adaptation to axial motor characteristics: The algorithm parameters of traditional control boards (such as magnetic field models and torque calculation logic) are based on radial motor design and do not match the characteristics of axial motors, such as "flat magnetic circuit and distributed windings". This results in insufficient motor efficiency, with the actual operating efficiency of axial motors being 8%-12% lower than the design value, and failing to fully reflect its energy-saving advantages. Summary of the Invention

[0008] Based on this, an axial motor control board, method, computer device, and storage medium are provided to solve the technical problem that existing control boards cannot meet the requirements of axial motors in terms of accuracy, real-time performance, reliability, and adaptability.

[0009] On the one hand, an axial motor control board is provided, which includes: a main control chip, a coprocessor, a communication interface module, a data acquisition module, a motor drive module, and a storage module;

[0010] The main control chip is interconnected with the coprocessor, and both the main control chip and the coprocessor are connected to the data acquisition module, the motor drive module and the storage module through the communication interface module.

[0011] The data acquisition module is used to collect the motor's three-phase current, bus voltage, winding temperature and speed data in real time;

[0012] The main control chip is used for system management, fault diagnosis, communication scheduling, and data storage;

[0013] The coprocessor is used to process real-time tasks of current decoupling, matrix calculation, and coordinate transformation of the field-oriented control (FOC) algorithm that controls the motor drive module.

[0014] Furthermore, the coprocessor includes:

[0015] The parameter identification module is used to identify the motor winding resistance and inductance parameters in real time based on the three-phase current, bus voltage, winding temperature and speed data of the motor collected in real time by the data acquisition module.

[0016] The FOC algorithm optimization module is used to decouple the current calculation based on the identified motor winding resistance and inductance parameters using the radix-2 fast Fourier transform (FFT) algorithm, and dynamically adjust the current loop and speed loop parameters of the field-oriented control (FOC) algorithm through the adaptive extended Kalman filter (AEKF) algorithm, and perform matrix calculation and coordinate transformation.

[0017] The control center is used to monitor, analyze, and record faults in real time the data acquisition module, the parameter identification module, and the FOC algorithm optimization module. It combines the three-phase current, bus voltage, winding temperature, and speed data of the motor to determine the motor's operating status and issue protection commands.

[0018] Furthermore, the main control chip includes:

[0019] The multi-parameter input module is used to acquire the motor's three-phase current, bus voltage, winding temperature, speed data, position data, communication status, pulse width modulation status, and power supply status through the communication interface module.

[0020] The Deep Neural Network (DNN) fault diagnosis model has an input layer, a hidden layer, and an output layer. The input layer is used to input the three-phase current, bus voltage, winding temperature, speed data, position data, communication status, pulse width modulation status, and power supply status of the motor. The hidden layer extracts fault features through the ReLU activation function. The output layer outputs winding short circuit, permanent magnet demagnetization, encoder fault, and controller fault.

[0021] The three-level protection mechanism module is used to perform actions such as cutting off pulse width modulation output (stopping motor drive), starting backup power supply, and sending fault report to coprocessor based on the output information of the multi-parameter input module and the fault output module.

[0022] Furthermore, the substrate of the axial motor control board is made of aluminum-based copper-clad laminate, and the thermal conductivity of the aluminum-based copper-clad laminate is greater than or equal to 200W / (m·K). The outer shell of the axial motor control board is made of magnesium-aluminum alloy shell with micro-arc oxidation treatment. The magnesium-aluminum alloy shell is provided with a serpentine heat dissipation channel, and a heat pipe is embedded in the serpentine heat dissipation channel. The diameter of the heat pipe is 3mm.

[0023] Furthermore, the data acquisition module includes a current sensor, a voltage sensor, a temperature sensor, and a position sensor; the communication interface module integrates multiple standard interfaces, including an encoder interface, a brake interface, and a debugging and software upgrade interface; the storage module contains a motor parameter database and a fast adaptation mechanism module. The motor parameter database contains magnetic field models, torque curve parameters, and preset control parameters for axial motors with power ratings from 50 to 500W. The fast adaptation mechanism module switches motor models via DIP switches or time-sensitive network communication; when a new motor is added, the motor ID is scanned via the local area network bus and the preset control parameters are loaded from the motor parameter database.

[0024] Furthermore, the main control chip also includes:

[0025] The status monitoring module is used to monitor the status data of the axial motor control board and the motor in real time. The status data of the axial motor control board includes chip temperature, power supply voltage, and communication link. The status data of the motor includes torque, speed, and winding temperature.

[0026] The data feedback module is used to upload the status data of the axial motor control board and the status data of the motor monitored and acquired by the status monitoring module to the control center of the coprocessor through time-sensitive network communication, and to perform data updates, real-time status monitoring and historical data recording.

[0027] The software upgrade module is used to perform online or offline upgrades based on the feedback data from the data feedback module, and to keep the motor in a braking state during the upgrade process.

[0028] Furthermore, the main control chip also includes:

[0029] The data analysis module is used to perform trend prediction, performance evaluation, fault warning and life prediction of the motor based on the status data of the axial motor control board and the status data of the motor monitored by the status monitoring module.

[0030] The control optimization module is used to adjust parameters based on the data from the data analysis module for trend prediction, performance evaluation, fault warning, and life prediction of the motor.

[0031] On the other hand, an axial motor control method is provided, including:

[0032] The data acquisition module collects real-time data on the motor's three-phase current, bus voltage, winding temperature, and speed.

[0033] The motor parameters are identified in real time by a coprocessor and processed in real time using the field-oriented control (FOC) algorithm for current decoupling, matrix calculation and coordinate transformation.

[0034] The main control chip uses a deep neural network (DNN) model to monitor motor faults in real time. If a fault is detected, a three-level protection mechanism is triggered to cut off the motor drive and send a fault report.

[0035] In another aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of an axial motor control method.

[0036] In another aspect, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of an axial motor control method.

[0037] The aforementioned axial motor control board, method, computer equipment, and storage medium, through a dual-processing architecture of a main control chip and a coprocessor within the axial motor control board, and their collaborative connection via communication interface modules with the data acquisition module, motor drive module, and storage module respectively, achieve high-speed acquisition of motor operating parameters, real-time control calculation, and division of labor for system management tasks. The main control chip is responsible for system scheduling, fault diagnosis, and data management, ensuring the stability of the control board under multi-tasking operation; the coprocessor is dedicated to performing high real-time tasks such as current decoupling, matrix calculation, and coordinate transformation in the field-oriented control (FOC) algorithm, improving the dynamic response performance of the current loop and speed loop. Real-time acquisition of key parameters such as three-phase current, bus voltage, winding temperature, and speed via the data acquisition module enables precise motor state perception and closed-loop control. Simultaneously, the drive module and storage module work collaboratively through a unified communication framework, making control command transmission and operational data storage more reliable. Therefore, this technical solution effectively improves the control accuracy, response speed, and operational stability of the axial motor, and enhances the reliability and safety of the control system under complex operating conditions. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a diagram of the multi-core heterogeneous hardware architecture of the axial motor control device in one embodiment of this application;

[0040] Figure 2 This is a block diagram of a coprocessor constituting an FOC algorithm optimization and real-time parameter identification system in one embodiment of this application;

[0041] Figure 3 This is a block diagram of a coprocessor constituting an intelligent fault diagnosis and self-protection system in one embodiment of this application;

[0042] Figure 4 This is a structural block diagram of a high thermal conductivity design system in one embodiment of this application;

[0043] Figure 5 This is a structural block diagram of an integrated axial motor adapter design system in one embodiment of this application;

[0044] Figure 6 This is a structural block diagram of a status monitoring and data feedback system in one embodiment of this application;

[0045] Figure 7This is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] In one embodiment, such as Figure 1 As shown, an axial motor control board is provided, which includes: a main control chip, a coprocessor, a communication interface module, a data acquisition module, a motor drive module, and a storage module;

[0048] The main control chip is interconnected with the coprocessor, and both the main control chip and the coprocessor are connected to the data acquisition module, the motor drive module and the storage module through the communication interface module.

[0049] The data acquisition module is used to collect the motor's three-phase current, bus voltage, winding temperature and speed data in real time;

[0050] The main control chip is used for system management, fault diagnosis, communication scheduling, and data storage;

[0051] The coprocessor is used to process real-time tasks of current decoupling, matrix calculation, and coordinate transformation of the field-oriented control (FOC) algorithm that controls the motor drive module.

[0052] The system employs a heterogeneous architecture of "main control chip (such as ARM Cortex-A76) + coprocessor (FPGA)": the main control chip is responsible for system management, fault diagnosis and communication, while the coprocessor focuses on real-time tasks such as current decoupling and torque calculation of the FOC algorithm, with a computational efficiency more than 3 times higher than that of a single chip.

[0053] The communication interface module integrates Time Sensitive Network (TSN) communication and supports PTP Precision Time Protocol (synchronization accuracy ≤ 1 microsecond), replacing traditional Ethernet to achieve low-latency transmission of control commands and status data (latency ≤ 1 millisecond), adapting to the multi-joint collaborative control requirements of humanoid robots.

[0054] The data acquisition module is equipped with a 16-bit high-precision ADC (sampling rate ≥ 1MHz) to collect parameters such as motor three-phase current, bus voltage, and winding temperature in real time, with a sampling error ≤ 0.5%, providing an accurate data foundation for parameter identification and fault diagnosis.

[0055] The motor drive module is used to implement a pulse width modulation generator, gate driver, overcurrent protection, and undervoltage protection.

[0056] like Figure 2 As shown, the coprocessor includes:

[0057] The parameter identification module is used to identify the motor winding resistance and inductance parameters in real time based on the three-phase current, bus voltage, winding temperature and speed data of the motor collected in real time by the data acquisition module.

[0058] The FOC algorithm optimization module is used to decouple the current calculation based on the identified motor winding resistance and inductance parameters using the radix-2 fast Fourier transform (FFT) algorithm, and dynamically adjust the current loop and speed loop parameters of the field-oriented control (FOC) algorithm through the adaptive extended Kalman filter (AEKF) algorithm, and perform matrix calculation and coordinate transformation.

[0059] The control center is used to monitor, analyze, and record faults in real time the data acquisition module, the parameter identification module, and the FOC algorithm optimization module. It combines the three-phase current, bus voltage, winding temperature, and speed data of the motor to determine the motor's operating status and issue protection commands.

[0060] like Figure 2 As shown, the coprocessor constitutes the FOC algorithm optimization and real-time parameter identification system.

[0061] Among them, an adaptive extended Kalman filter (AEKF) algorithm is introduced: based on the current, voltage and speed data collected by the ADC, the motor winding resistance, inductance and other parameters are identified in real time (identification response time ≤100 microseconds), and the current loop and speed loop parameters of the FOC algorithm are dynamically adjusted to ensure control accuracy under load changes (0-100% rated load) and temperature fluctuations (-40℃-105℃).

[0062] The FOC algorithm optimization module optimizes the current decoupling operation: it adopts the radix-2 fast Fourier transform (FFT) algorithm to shorten the current decoupling operation time from the traditional 3-4 milliseconds to 1-2 milliseconds, improves the real-time performance of torque control, and makes the motor torque ripple rate ≤5%.

[0063] By incorporating a parameter identification module and a FOC algorithm optimization module within the coprocessor, the control board maintains stable control accuracy despite variations in motor load, temperature, and parameter drift. The parameter identification module identifies key parameters such as resistance and inductance in real time, preventing the performance degradation that traditional motors experience after temperature rise. Combining a radix-2 FFT-optimized current decoupling algorithm with the AEKF dynamic parameter adjustment method results in faster FOC algorithm calculations, enhanced real-time performance, significantly reduced motor torque ripple, and improved operational stability. Simultaneously, the control center provides unified management of all computational modules, ensuring system traceability and reliability.

[0064] like Figure 3 As shown, the main control chip includes:

[0065] The multi-parameter input module is used to acquire the motor's three-phase current, bus voltage, winding temperature, speed data, position data, communication status, pulse width modulation status, and power supply status through the communication interface module.

[0066] The Deep Neural Network (DNN) fault diagnosis model has an input layer, a hidden layer, and an output layer. The input layer is used to input the three-phase current, bus voltage, winding temperature, speed data, position data, communication status, pulse width modulation status, and power supply status of the motor. The hidden layer extracts fault features through the ReLU activation function. The output layer outputs winding short circuit, permanent magnet demagnetization, encoder fault, and controller fault.

[0067] The three-level protection mechanism module is used to perform actions such as cutting off pulse width modulation output (stopping motor drive), starting backup power supply, and sending fault report to coprocessor based on the output information of the multi-parameter input module and the fault output module.

[0068] By integrating a multi-parameter input module and a deep neural network (DNN) fault diagnosis model into the main control chip, the control board can identify complex motor faults faster and more accurately than traditional threshold detection. The DNN model can extract correlation features from multi-dimensional data such as current, voltage, and temperature, significantly improving fault identification accuracy and effectively identifying fault types that are difficult to detect with traditional systems, such as winding short circuits, demagnetization, and encoder signal abnormalities. A three-level protection mechanism ensures that the motor drive stops immediately, critical chips are not powered off, and information is uploaded in a timely manner when a fault occurs, thereby greatly improving system safety, especially suitable for high-safety scenarios such as medical surgical robots.

[0069] like Figure 3 As shown, the coprocessor constitutes an intelligent fault diagnosis and self-protection system.

[0070] Construct a deep neural network (DNN) fault diagnosis model: The input layer contains 8 parameters such as current, voltage, temperature, speed, and encoder signal. The hidden layer (3 layers) extracts fault features through the ReLU activation function. The output layer can identify 12 types of faults such as "winding short circuit", "permanent magnet demagnetization" and "encoder failure", with a false diagnosis rate of ≤5%.

[0071] Multiple self-protection mechanisms: When a fault occurs, the system triggers three levels of protection within 5 milliseconds. First, it cuts off the pulse width modulation output (stops motor drive), then it starts the backup power supply to maintain the core chip monitoring function, and finally it sends a fault report (including fault type, location, and cause) to the control center through time-sensitive network communication to avoid secondary damage.

[0072] like Figure 4 As shown, the substrate of the axial motor control board is made of aluminum-based copper-clad laminate, and the thermal conductivity of the aluminum-based copper-clad laminate is greater than or equal to 200W / (m·K). The outer shell of the axial motor control board is made of magnesium-aluminum alloy shell with micro-arc oxidation treatment. The magnesium-aluminum alloy shell is provided with a serpentine heat dissipation channel, and a heat pipe is embedded in the serpentine heat dissipation channel. The diameter of the heat pipe is 3mm.

[0073] like Figure 4 As shown, the substrate uses a high thermal conductivity aluminum-based copper-clad laminate (thermal conductivity ≥200W / (m·K)), replacing the traditional FR4 substrate (thermal conductivity ≤0.3W / (m·K)), improving the heat dissipation efficiency of the core chip area by more than 60%. The control board shell is made of magnesium-aluminum alloy with micro-arc oxidation treatment (thermal conductivity improved by 40%-60%), and a serpentine heat dissipation channel is designed: ultra-thin heat pipes (3mm in diameter) are embedded in the channel to quickly conduct the heat of the core chip to the shell, and the chip temperature can be controlled below 75℃ under high temperature environment (105℃).

[0074] The system employs a high thermal conductivity aluminum-based copper-clad laminate, a magnesium-aluminum alloy casing, and a serpentine heat dissipation channel and heat pipe structure to create a highly efficient temperature control system. The high thermal conductivity substrate allows for rapid heat dissipation from the power devices in the motor drive, preventing localized hotspots. The magnesium-aluminum alloy casing further enhances heat dissipation efficiency. The serpentine channel and 3mm heat pipes achieve rapid temperature reduction of the core chip through phase change heat transfer. This comprehensive heat dissipation structure enables stable operation under high-temperature conditions, preventing control errors or system frequency throttling caused by overheating, and allowing the motor to maintain high-performance output for extended periods.

[0075] Furthermore, the data acquisition module includes a current sensor, a voltage sensor, a temperature sensor, and a position sensor; the communication interface module integrates multiple standard interfaces, including an encoder interface, a brake interface, and a debugging and software upgrade interface; the storage module contains a motor parameter database and a fast adaptation mechanism module. The motor parameter database contains magnetic field models, torque curve parameters, and preset control parameters for axial motors with power ratings from 50 to 500W. The fast adaptation mechanism module switches motor models via DIP switches or time-sensitive network communication; when a new motor is added, the motor ID is scanned via the local area network bus and the preset control parameters are loaded from the motor parameter database.

[0076] The data acquisition module incorporates multiple sensors, including those for current, voltage, temperature, and position, enabling it to provide comprehensive and accurate operational information and improving the input accuracy of the FOC algorithm and fault diagnosis model. The communication interface module integrates various standard interfaces, ensuring the control board is compatible with multiple encoder and brake models without external adapters, reducing system size and wiring complexity. The motor parameter database and rapid adaptation mechanism in the storage module eliminate the need for recalibration when replacing or adding motor models, significantly shortening the debugging cycle and improving the control board's versatility and engineering application efficiency.

[0077] like Figure 5 As shown, the communication interface module integrates multiple standard interfaces: including RS485 (encoder interface, supporting Gray code / binary code absolute encoders), CANopen (brake interface, adapted for electromagnetic brakes), and USB-Type-C (debugging and software upgrade), eliminating the need for additional adapter modules and reducing axial dimensions by 15%.

[0078] The storage module contains a motor parameter database: it includes magnetic field models and torque curve parameters for axial motors with power ratings from 50 to 500W. Motor models can be quickly switched via DIP switches or time-sensitive network communication, shortening the adaptation cycle to one week. When a new motor is added, the control board can automatically scan the motor ID and load the default control strategy via the CAN bus, eliminating the need for manual configuration.

[0079] like Figure 6 As shown, the main control chip also includes:

[0080] The status monitoring module is used to monitor the status data of the axial motor control board and the motor in real time. The status data of the axial motor control board includes chip temperature, power supply voltage, and communication link. The status data of the motor includes torque, speed, and winding temperature.

[0081] The data feedback module is used to upload the status data of the axial motor control board and the status data of the motor monitored and acquired by the status monitoring module to the control center of the coprocessor through time-sensitive network communication, and to perform data updates, real-time status monitoring and historical data recording.

[0082] The software upgrade module is used to perform online or offline upgrades based on the feedback data from the data feedback module, and to keep the motor in a braking state during the upgrade process.

[0083] The control board features real-time health monitoring capabilities through a status monitoring module and a data feedback module. This allows for continuous monitoring of key indicators such as chip temperature, motor temperature, and communication status during operation. Time-sensitive network communication ensures high-bandwidth, low-latency data upload, improving the monitoring accuracy of the host computer. The software upgrade module supports both online and offline upgrades while ensuring the motor remains braked during the upgrade process, enhancing the safety and flexibility of system maintenance. This makes it suitable for intelligent motor systems requiring continuous algorithm updates.

[0084] Furthermore, the main control chip also includes:

[0085] The data analysis module is used to perform trend prediction, performance evaluation, fault warning and life prediction of the motor based on the status data of the axial motor control board and the status data of the motor monitored by the status monitoring module.

[0086] The control optimization module is used to adjust parameters based on the data from the data analysis module for trend prediction, performance evaluation, fault warning, and life prediction of the motor.

[0087] By incorporating data analysis and control optimization modules, the motor control system is upgraded from "passive control" to "predictive optimization control." The data analysis module can predict future motor trends, warn of potential faults, and assess motor performance degradation based on historical and real-time data, allowing for proactive measures. The control optimization module automatically adjusts control parameters based on the analysis results, ensuring the motor maintains high efficiency, low energy consumption, and high precision during long-term operation. This feature significantly improves the motor's reliability and lifespan.

[0088] like Figure 6 As shown, the system monitors the control board's own status (chip temperature, power supply voltage, communication link) and the motor status (torque, speed, winding temperature) in real time, and uploads the data to the control center via Time-Sensitive Network (TSN) communication, with a data update frequency ≥100Hz. It supports dual-bus software upgrades: the control algorithm and fault diagnosis model can be updated via Time-Sensitive Network (TSN) bus (online upgrade) or USB-Type-C (offline upgrade), without interrupting the motor's basic operation (e.g., maintaining braking state), adapting to later functional iterations.

[0089] Compared with traditional control boards, the intelligent control board for axial motors in this application has the following significant advantages:

[0090] 1. Significantly improved control performance: Through a multi-core heterogeneous architecture and AEKF algorithm, motor parameters are identified in real time, FOC control accuracy reaches ±0.05-0.1mm, and command response latency is ≤1 millisecond, meeting the needs of precision scenarios such as humanoid robot joints and medical surgical robots; torque ripple rate is ≤5%, and motor operation stability is improved by more than 30%.

[0091] 2. Significantly enhanced fault safety: The DNN fault diagnosis model can accurately identify 12 types of faults with a false diagnosis rate of ≤5%, which is 80% lower than traditional threshold detection; it triggers three-level protection within 5 milliseconds to avoid motor damage, and improves reliability by 60% in safety-sensitive scenarios such as medical and industrial applications.

[0092] 3. Heat dissipation capacity adapted to high power requirements: The combination design of high thermal conductivity aluminum-based copper-clad laminate, magnesium-aluminum alloy shell and serpentine heat pipe improves heat dissipation efficiency by 40%-60%. The chip temperature is ≤75℃ in a high temperature environment of 105℃, which solves the overheating problem of axial motors when operating at high power density and extends the continuous running time of the motor by more than 2 times.

[0093] 4. Integration and compatibility optimization: Integrated encoder and brake standard interfaces reduce axial dimensions by 15%; compatible with 50-500W axial motors, shortening the adaptation cycle from 2-3 months to 1 week, reducing the cost and time cost of motor replacement for enterprises; automatic scanning of new motors and loading strategies improves system scalability by 50%.

[0094] 5. Improved adaptability of axial motor characteristics: Optimized magnetic field model and torque control logic improve the actual operating efficiency of axial motor by 8%-12% and reduce energy consumption by 15%; supports dual-bus software upgrades, and no hardware replacement is required for later function iterations, extending the service life of the control board (from 3 to 5 years).

[0095] Wide applicability: It can be adapted to different scenarios such as joint motors of humanoid robots, drive motors of medical equipment, and auxiliary motors of new energy vehicles. It also supports multi-motor collaborative control through time-sensitive network communication, providing core support for the intelligentization of high-end equipment.

[0096] On the other hand, an axial motor control method is provided, including:

[0097] The data acquisition module collects real-time data on the motor's three-phase current, bus voltage, winding temperature, and speed.

[0098] The motor parameters are identified in real time by a coprocessor and processed in real time using the field-oriented control (FOC) algorithm for current decoupling, matrix calculation and coordinate transformation.

[0099] The main control chip uses a deep neural network (DNN) model to monitor motor faults in real time. If a fault is detected, a three-level protection mechanism is triggered to cut off the motor drive and send a fault report.

[0100] The axial motor control method integrates data acquisition, parameter identification, algorithm optimization, fault protection, heat dissipation management, and monitoring feedback into a complete closed-loop control process, enabling the system to adapt to changes in the operating environment and maintain highly accurate control performance. By dynamically adjusting FOC control parameters and automatically triggering fault protection mechanisms, the system ensures stability and safety even under harsh conditions such as high speed, high temperature, and high load. Furthermore, online upgrade capabilities ensure the system's advantages of continuous iteration and long-term stable use.

[0101] Furthermore, before performing current decoupling, matrix calculation, and coordinate transformation, the coprocessor also uses an adaptive extended Kalman filter algorithm to estimate the motor winding resistance and inductance parameters online to compensate for motor parameter drift caused by temperature rise and load changes, thereby improving the robustness of field-oriented control.

[0102] Furthermore, the coprocessor employs a radix-2 fast Fourier transform structure for computational optimization when performing coordinate transformation and current decoupling operations, thereby reducing the computational power consumption of the coprocessor, improving real-time computation efficiency, and expanding the control bandwidth of the current loop and speed loop.

[0103] Furthermore, the deep neural network fault diagnosis model of the main control chip is based on multi-dimensional input data, including current vector fluctuation characteristics, bus voltage transient change characteristics, winding temperature rise trend characteristics, and encoder feedback consistency characteristics, to perform inference, thereby achieving accurate identification of winding short circuits, permanent magnet demagnetization, encoder faults, and controller anomalies.

[0104] Furthermore, the three-level protection mechanism includes:

[0105] (1) Level 1 protection: When a potential abnormality is detected, the pulse width modulation (PWM) duty cycle is automatically reduced and the output torque is limited to avoid sudden current surges;

[0106] (2) Secondary protection: When the fault worsens, the PWM output of the motor drive module is immediately cut off to prevent damage to the power devices;

[0107] (3) Level 3 protection: When a serious fault is confirmed, the motor brake is triggered to lock the motor rotor, and a fault report and related fault data are sent to the coprocessor via the TSN bus.

[0108] Furthermore, when the three-level protection mechanism is triggered, after the fault is eliminated or the abnormal state is confirmed to be resolved, the main control chip automatically executes the recovery process. The recovery process includes zero-point position recalibration, motor parameter model reloading, and control loop parameter reinitialization, thereby achieving safe recovery of the motor drive.

[0109] Furthermore, the method also includes storing the motor operating parameters, control command data, real-time identification results, and fault diagnosis results calculated in each cycle into a non-volatile memory to form traceable operating history data, providing a basis for subsequent system maintenance, performance optimization, and life assessment.

[0110] Furthermore, the method also includes: automatically adjusting the heat dissipation strategy of the control board based on the temperature of the main control chip, the temperature of the power module and the temperature of the motor winding collected by the data acquisition module, including starting or increasing the working power of the active heat dissipation device and enabling the heat pipe auxiliary heat dissipation path, so as to ensure that the control board and the motor operate stably under high load conditions.

[0111] Furthermore, when the deep neural network model detects a potential fault risk but does not reach the fault judgment threshold, the main control chip performs risk avoidance control on the motor in advance, including reducing the maximum output torque, limiting the maximum speed, or adjusting the current loop bandwidth, in order to reduce potential damage and extend the service life of the motor and control board.

[0112] Taking the application scenario of "axial joint motor of medical surgical robot" as an example, the specific implementation process of the present invention is explained as follows:

[0113] First, application scenario requirements: The end joint of the surgical robot uses a 50W axial motor, which needs to achieve sub-millimeter positioning accuracy (±0.08mm), fast response (command delay ≤1 millisecond), and avoid motion deviation caused by motor failure (such as short circuit of winding) during the operation. At the same time, it needs to be able to adapt to the high temperature environment of the operating room (30℃-40℃).

[0114] Second, control board configuration and adaptation:

[0115] Hardware connections: The control board connects to a 16-bit absolute Gray code encoder via an RS485 interface (to obtain motor position), to an electromagnetic brake via a CANopen interface (to keep the joint locked when the operation is paused), and communicates with the robot control center via time-sensitive network communication.

[0116] Motor parameter adaptation: Select the "50W axial motor" model via the DIP switch, and the control board will automatically load the pre-stored magnetic field model (adapting to the axial magnetic flux distribution) and torque curve. No manual adjustment is required, and the adaptation time is about 5 minutes.

[0117] Third, the operation process:

[0118] During the surgery, the FPGA coprocessor on the control board controls the motor torque in real time through the FOC algorithm: the ADC collects the three-phase current and winding temperature at a sampling rate of 1MHz, the AEKF algorithm identifies the winding resistance in real time (due to the long operation time, the temperature rises from 30℃ to 50℃, and the resistance increases by 8%), and dynamically adjusts the current loop parameters to ensure torque control accuracy, with the robot end-effector positioning error ≤0.08mm.

[0119] If the motor experiences an "encoder signal loss" fault (such as poor cable contact), the DNN model can identify the fault within 0.5 milliseconds and trigger protection within 5 milliseconds by comparing the characteristics of missing speed feedback and abnormal current fluctuations: cutting off the pulse width modulation output (stopping the motor drive), activating the brake to lock the joint, and sending an "encoder fault" report to the control center via time-sensitive network communication. Doctors can then promptly suspend the surgery to avoid surgical errors.

[0120] When the operating room temperature is 35℃, the aluminum-based copper-clad board of the control board conducts the heat of the core chip (FPGA) to the magnesium-aluminum alloy shell. The serpentine heat pipe accelerates heat dissipation, and the chip temperature is stable at 65℃ without overheat protection triggering. The motor runs continuously for 4 hours without performance fluctuation.

[0121] The intelligent control board for axial motors in this application has significant advantages thanks to multiple core technologies: the hardware adopts a multi-core heterogeneous architecture, coupled with a dedicated communication module and high-precision parameter acquisition components, resulting in superior computing efficiency, command response speed, and data acquisition accuracy compared to traditional single-chip designs; the algorithm optimizes the FOC control logic and enables real-time identification of motor parameters, accurately adapting to the characteristics of axial motors and improving control accuracy and motor operating efficiency; the fault protection end constructs a multi-parameter intelligent diagnostic model, which can accurately identify various faults and trigger a graded protection mechanism, with reliability far exceeding traditional threshold detection methods; the heat dissipation design uses high thermal conductivity materials and an efficient heat dissipation structure, ensuring stable operation in high-temperature environments; in terms of integrated adaptation, it has built-in multiple standard interfaces that eliminate the need for additional adapter modules and is compatible with axial motors of different power, significantly shortening the adaptation cycle.

[0122] In the aforementioned axial motor control method, a dual-processing architecture of a main control chip and a coprocessor is implemented in the axial motor control board. These two chips are connected to the data acquisition module, motor drive module, and storage module via communication interface modules, respectively. This achieves a division of labor for high-speed acquisition of motor operating parameters, real-time control calculations, and system management tasks. The main control chip is responsible for system scheduling, fault diagnosis, and data management, ensuring the stability of the control board under multi-tasking operation. The coprocessor is dedicated to performing high real-time tasks in the Field-Oriented Control (FOC) algorithm, such as current decoupling, matrix calculation, and coordinate transformation, improving the dynamic response performance of the current and speed loops. The data acquisition module acquires key parameters such as three-phase current, bus voltage, winding temperature, and speed in real time, enabling precise motor state perception and closed-loop control. Simultaneously, the drive module and storage module work collaboratively through a unified communication framework, making control command transmission and operational data storage more reliable. Therefore, this technical solution effectively improves the control accuracy, response speed, and operational stability of the axial motor, and enhances the reliability and safety of the control system under complex operating conditions.

[0123] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores axial motor control data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements an axial motor control method.

[0124] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0125] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of an axial motor control method.

[0126] For specific limitations on the steps implemented by the processor when executing a computer program, please refer to the limitations on the method of axial motor control mentioned above, which will not be repeated here.

[0127] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of an axial motor control method.

[0128] For specific limitations on the implementation steps when a computer program is executed by a processor, please refer to the limitations on the method of axial motor control mentioned above, which will not be repeated here.

[0129] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0130] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0131] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An axial motor control board, characterized in that, include: Main control chip, coprocessor, communication interface module, data acquisition module, motor drive module, and storage module; The main control chip is interconnected with the coprocessor, and both the main control chip and the coprocessor are connected to the data acquisition module, the motor drive module and the storage module through the communication interface module. The data acquisition module is used to collect the motor's three-phase current, bus voltage, winding temperature and speed data in real time; The main control chip is used for system management, fault diagnosis, communication scheduling, and data storage; The coprocessor is used to process the real-time tasks of current decoupling, matrix calculation and coordinate transformation of the field-oriented control algorithm that controls the motor drive module. The coprocessor includes: The parameter identification module is used to identify the motor winding resistance and inductance parameters in real time based on the three-phase current, bus voltage, winding temperature and speed data of the motor collected in real time by the data acquisition module. The FOC algorithm optimization module is used to employ the radix-2 fast Fourier transform algorithm to decouple the current based on the identified motor winding resistance and inductance parameters, and dynamically adjust the current loop and speed loop parameters of the field-oriented control algorithm through the adaptive extended Kalman filter algorithm, and perform matrix calculations and coordinate transformations. The control center is used to monitor, analyze, and record faults in real time the data acquisition module, the parameter identification module, and the FOC algorithm optimization module. It combines the three-phase current, bus voltage, winding temperature, and speed data of the motor to determine the motor's operating status and issue protection commands.

2. The axial motor control board according to claim 1, characterized in that, The main control chip includes: The multi-parameter input module is used to acquire the motor's three-phase current, bus voltage, winding temperature, speed data, position data, communication status, pulse width modulation status, and power supply status through the communication interface module. The deep neural network fault diagnosis model has an input layer, a hidden layer, and an output layer. The input layer is used to input the three-phase current, bus voltage, winding temperature, speed data, position data, communication status, pulse width modulation status, and power supply status of the motor. The hidden layer extracts fault features through the ReLU activation function. The output layer outputs winding short circuit, permanent magnet demagnetization, encoder fault, and controller fault. The three-level protection mechanism module is used to perform actions such as cutting off pulse width modulation output, starting backup power supply, and sending fault report to the coprocessor based on the output information of the multi-parameter input module and the fault output module.

3. The axial motor control board according to claim 1, characterized in that, The substrate of the axial motor control board is made of aluminum-based copper-clad laminate with a thermal conductivity greater than or equal to 200 W / (m·K). The outer shell of the axial motor control board is made of magnesium-aluminum alloy shell with micro-arc oxidation treatment. The magnesium-aluminum alloy shell is provided with a serpentine heat dissipation channel, and a heat pipe is embedded in the serpentine heat dissipation channel with a diameter of 3 mm.

4. The axial motor control board according to claim 1, characterized in that, The data acquisition module includes a current sensor, a voltage sensor, a temperature sensor, and a position sensor; the communication interface module integrates multiple standard interfaces, including an encoder interface, a brake interface, and a debugging and software upgrade interface; the storage module contains a motor parameter database and a fast adaptation mechanism module. The motor parameter database contains magnetic field models, torque curve parameters, and preset control parameters for axial motors with power ratings from 50 to 500W. The fast adaptation mechanism module switches motor models via DIP switches or time-sensitive network communication; when a new motor is added, the motor ID is scanned via the local area network bus and the preset control parameters are loaded from the motor parameter database.

5. The axial motor control board according to claim 1, characterized in that, The main control chip also includes: The status monitoring module is used to monitor the status data of the axial motor control board and the motor in real time. The status data of the axial motor control board includes chip temperature, power supply voltage, and communication link. The status data of the motor includes torque, speed, and winding temperature. The data feedback module is used to upload the status data of the axial motor control board and the status data of the motor monitored and acquired by the status monitoring module to the control center of the coprocessor through time-sensitive network communication, and to perform data updates, real-time status monitoring and historical data recording. The software upgrade module is used to perform online or offline upgrades based on the feedback data from the data feedback module, and to keep the motor in a braking state during the upgrade process.

6. The axial motor control board according to claim 5, characterized in that, The main control chip also includes: The data analysis module is used to perform trend prediction, performance evaluation, fault warning and life prediction of the motor based on the status data of the axial motor control board and the status data of the motor monitored by the status monitoring module. The control optimization module is used to adjust parameters based on the data from the data analysis module for trend prediction, performance evaluation, fault warning, and life prediction of the motor.

7. An axial motor control method, characterized in that, For an axial motor control board according to any one of claims 1 to 6, the method comprises: The data acquisition module collects real-time data on the motor's three-phase current, bus voltage, winding temperature, and speed. The motor parameters are identified in real time by a coprocessor and processed in real time using the current decoupling, matrix calculation and coordinate transformation tasks of the field-oriented control algorithm. The main control chip uses a deep neural network model to monitor motor faults in real time. If a fault is detected, a three-level protection mechanism is triggered to cut off the motor drive and send a fault report.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method of claim 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method of claim 7.

Images